This page presents the extensive results list of the paper entitled "Data Poisoning: When ML feels ill".
The results are divided according to the mitigation system implemented: No voting system and Voting system with 3, 5, or 7 votes.
To see each of this systems please pick one in the dropdown bellow.

Overall analysis

Random attack

Switch attack

Flip attack

Overall analysis

Random attack

Switch attack

Flip attack

Overall analysis

Random attack

Switch attack

Flip attack

Individual analysis - Random

Logistic Regression

Support Vector Machine

K- Nearest Neighbors (1)

K- Nearest Neighbors (3)

K- Nearest Neighbors (5)

K- Nearest Neighbors (7)

K- Nearest Neighbors (9)

Decision Tree

Artificial Neural Network

Random Forest

Voting Classifier (hard)

Voting Classifier (soft)

Individual analysis - Switch

Logistic Regression

Support Vector Machine

K- Nearest Neighbors (1)

K- Nearest Neighbors (3)

K- Nearest Neighbors (5)

K- Nearest Neighbors (7)

K- Nearest Neighbors (9)

Decision Tree

Artificial Neural Network

Random Forest

Voting Classifier (hard)

Voting Classifier (soft)

Individual analysis - Flip

Logistic Regression

Support Vector Machine

K- Nearest Neighbors (1)

K- Nearest Neighbors (3)

K- Nearest Neighbors (5)

K- Nearest Neighbors (7)

K- Nearest Neighbors (9)

Decision Tree

Artificial Neural Network

Random Forest

Voting Classifier (hard)

Voting Classifier (soft)